• Please review our updated Terms and Rules here

If you wanted an Emulator what would it be?

Cool. If you dug into it, you may have noticed the bootstrap process was a nightmare.
Very much a one-off prototype.
 
rcman2 said:
That's funny you should say that, I wanted to make one. I already made a Xenix and VAX/VMS one for the PICO in MicroPython.

RC
Click to expand...
Oh, sorry, I mixed things up.
I was thinking about the AS/400 which I think there aren't any emulators for.
(or for that sake any of the related systems, I think they are known as "IBM i" or something similar).
 
rcman2 said:
I did start an emulator for the BCC-500

github.com

clones/BCC500 at main · rcman/clones

Contribute to rcman/clones development by creating an account on GitHub.
github.com github.com

RC
Click to expand...
You should know that much of the operating system was written in microcode. Most of the processors were dedicated to the system. Two processors were for running user code, which could be the native BCC 500 instruction set, or emulating the SDS 940 (I know you knew this part). The emulator I started is for running the microcode.

The BCC 500 was an ARPANET host. My motivation flagged when I didn't see any networking code in the avilable software. It seems to me there was a PDP-11 ELF frontend handling the network.
 
"A multi-processor emulator for the fictional BCC-500 computer system"

The system was not "fictional"
It was built in Berkeley and Wayne Lichtenberger moved to the University of Hawaii and
reassembled it there
http://bitsavers.org/pdf/bcc the original installation
http://bitsavers.org/pdf/univOfHawaii/Aloha_BCC-500 the re-installation

Wayne saved a massive amount of documentation and tapes, which were donated
to the Computer History Museum when he passed recently. I went up to Seattle in
the 00s and read the 9-track tapes. There are 7-track tapes from the SDS-940, which
Wayne also was involved in at Cal, that I haven't read yet.
 

Attachments

  • BCC-500_1.jpg
    BCC-500_1.jpg
    1.2 MB · Views: 8
  • hawaii_bcc500_block_diagram.jpg
    hawaii_bcc500_block_diagram.jpg
    503.8 KB · Views: 7
Al Kossow said:
you can see the PDP11/10 in the block diagram that I just posted
Click to expand...
The emulator emulates the hardware, tape drive, card reader and console.

I’ve read a few things that say it didn’t act alone when it was in Hawaii but I also read it was the father of Ethernet.

The emulator try’s to allow you to compile code. I’ll need see if I can find some besides what’s in the emulator.

It’s a starting point, nothing really out there now. So let’s just go with what we have.

Rc
 
I know there are several excellent PC emulators around but so far none of them emulates the V20/V30 CPU including its 8080 mode.
 
Ruud said:
I know there are several excellent PC emulators around but so far none of them emulates the V20/V30 CPU including its 8080 mode.
Click to expand...
You'll need to explain that a little more to me. I did see someone made an emulator for a CPU of that era but I just need to understand the difference.
Thanks
RC
 
rcman2 said:
You'll need to explain that a little more to me. I did see someone made an emulator for a CPU of that era but I just need to understand the difference.
Thanks
RC
Click to expand...
I'm not Ruud, but the gist of it is that the V20 have a mode where it can run 8080 code, and thus by replacing the 8088 by a V20 in a regular PC it's possible to run 8-bit 8088 CP/M programs, which seems to have been a thing for a while in the 80's.
 
So update to the binary extractor for the BCC-500. It seems to work and report back. Now its just a question of what to do with that data besides trying to run it.

BinaryExtractor.png
 
I need to push the full set of tapes to bitsavers.
The "proof tape" is like a tar file, it is not executable and the .pdf in that directory describes the format
It is not a byte stream, record marks matter.
I need to dig back out the work I did in the 00s to try to understand what was on the tapes.
I scanned a lot of Wayne's documentation back then, but not everything, in particular the changes
done in Hawaii. I have everything scanned now after CHM received his files from his estate in
2024.

Assembly_of_the_Initial_Core_Load.pdf is what you need to get the actual native machine started.
The real computer instruction set is MUCH more complicated than what you have implemented in the 940 emulation
The bootstrap process involved the HP2100, which controlled the tape drive loading the operating system

The system documentation doesn't make a lot of sense. You need to look at all of the documents in the bcc and univOfHawaii
directories. It is also confusing because the original and rebuilt system documents are intermixed. Some parts of the system
like the file system processor are only described in bcc documents and in Mel Pertile's thesis. They made extensive mods
in Hawaii to extend the core memory. Actually, it isn't clear how much of that work was ever finished before the system was
retired in the late 70s.

Since they also worked on building the Project Genie SDS940 system at Cal, there are dump tapes from that system mixed
in with the BCC system. The point of the 940 emulation was to bootstrap the BCC using code they wrote at Cal. Some of
that bootstrapping work was done on Shell Research's SDS 940 system in Berkeley.

also, FYI, the 940 system used SDS character encoding, not ASCII

I don't have the processor microcode listings on line either, I notice.
I think the point of putting the proof tape up was that the running microcode files are on the tape.
The microcode was changed a bit, but not much. It required removing or soldering in diodes for
each microcode instruction bit. CHM has at least some of those boards but I've never photographed
them. https://www.computerhistory.org/collections/catalog/102728905/ for example
 
Last edited:
Al Kossow said:
I need to push the full set of tapes to bitsavers.
The "proof tape" is like a tar file, it is not executable and the .pdf in that directory describes the format
It is not a byte stream, record marks matter.

Assembly_of_the_Initial_Core_Load.pdf is what you need to get the actual native machine started.
The real computer instruction set is MUCH more complicated than what you have implemented in the 940 emulation
The bootstrap process involved the HP2100, which controlled the tape drive loading the operating system

The system documentation doesn't make a lot of sense. You need to look at all of the documents in the bcc and univOfHawaii
directories. It is also confusing because the original and rebuilt system documents are intermixed. Some parts of the system
like the disk controller processor are only described in bcc documents and in Mel Pertile's thesis. They made extensive mods
in Hawaii to extend the core memory. Actually, it isn't clear how much of that work was ever finished before the system was
retired in the late 70s.

Since they also worked on building the Project Genie SDS940 system at Cal, there are dump tapes from that system mixed
in with the BCC system. The point of the 940 emulation was to bootstrap the BCC using code they wrote at Cal
Click to expand...
Check this file.

github.com

clones/BCC-Emulator_with_GUI/Initial_Core_load.md at main · rcman/clones

Contribute to rcman/clones development by creating an account on GitHub.
github.com github.com
 
Sorry I need the super brains here. I passed what the pdf said through what I have now and this is WAY Bigger than I thought. I need people to review this document please.

github.com

clones/BCC-Emulator_with_GUI/problem_with_boot.md at main · rcman/clones

Contribute to rcman/clones development by creating an account on GitHub.
github.com github.com

Please and thanks
The main takeaway is this:


The biggest question: Does the BCC-500 architecture match the SDS 940 architecture described here? If they’re significantly different, you may need to create an adapted startup sequence rather than directly implementing this SDS 940 ICL approach.


Edit.

I have the answer. They are COMPLETELY different. Houston we have a problem…..
 
Last edited:
Okay never being happy with that I did some digging. I found documentation right from
The ALOHA System
F.F. Kuo
University of Hawaii
(Originally Published in Vol. 4, No. 1, January 1974)

AlohaSystem.png
 
rcman2 said:
You'll need to explain that a little more to me. I did see someone made an emulator for a CPU of that era but I just need to understand the difference.
Thanks
RC
Click to expand...
The NEC V20 can be seen as an Intel 8088 and an Intel 8080 in one processor. The V30 is 8086 plus 8080. CP/M was originally developed for the 8080. The V20-MBC proofs that it is possible to run 8086 code and CP/M in the same computer because of the use of the V20. Run, not emulating it! If this V20-MBC can do it, why not any PC equipped with a V20/V30?
 
You must log in or register to reply here.
Back
Top